Project Report

Permanent URI for this community

http://172.16.1.132:4000/handle/123456789/5368

Browse

Now showing 1 - 20 of 1079

3D Model Generation
(2015-12-17T10:40:11Z) Aalhad, Bhake; Gurjinder Pal, Singh; Vatan, Yadav; Sachpreet, Singh
Three Dimensional or 3D is referring to an object which has length, width and depth, in the other word all the object in the real world are exist in 3D form. Since past decade of years, 3D technology remains one of the hottest technologies that many people are discussing and studying wisely all around the world.3D technology is a complete resource for professionals in designing area which interested in CAD, modeling and rendering software. Computer vision deals with automatic extraction of various kinds of information from images. Various algorithms have been discussed for reconstruction purpose. Present research gives an overview of some of these algorithms and then it discusses the implementation issues of one of that algorithm. Volumetric reconstruction algorithms are getting popular due to their less complexity and the increasing storing capability of the computers. This project focuses on how the voxel occupy algorithm behaves. The project consists of two parts which are the software part and the hardware part. The first step of the procedure for the project is the software part. Initially, the 2D image is captured from camera calibration and data image before obtaining the silhouette of the image. The combination of finding model, make voxel and carving voxel in order to get the surface of the voxel. Lastly, the project colored the surface of the model and displays the finishing model. The successful developed coding enables to display the 3D image.
A Compact Speech-Driven Framework for Multilingual Voice Exchange Using Indian Languages
(2026-02-04) G Ramana Gowda Patil 1NH22CS073; Manjunatha Royal J 1NH22CS090; Yashwanth Reddy K V 1NH22CS101; K S Sharmila 1NH22CS109
The increasing demand for real-time multilingual communication has highlighted the need for speech systems that are lightweight, efficient and capable of delivering natural and accurate outputs. Existing solutions often rely on heavy computational resources and struggle with preserving tone, clarity and expressiveness, which limits their usability in practical environments. To address these challenges, this work introduces INDRA-Lite, a compact three stage framework designed for seamless speech-to speech translation across regional languages. The system incorporates a phoneme-aware speech recognition module that captures linguistic tones and acoustic variations, a bilingual mapping engine built on a gated recurrent architecture with language tokens for precise alignment and a prosody-driven synthesis unit optimized to maintain intonation and clarity in the generated speech.
A Deep learning model for collective disorder using Visual Geometry Group 16
(NHCE, 2022) V RAVI RAJ : UJWAL V : SACHETH N K
In the current scenario, people have become vulnerable to various diseases due to their lifestyle and their environment. Many of the analyses that have already been done looked at specific diseases. A user needs to use one analysis when they want to analyse diabetes, and another analysis when they want to analyse heart disease. This process takes a while. Moreover, if any user having multiple diseases, but the current method can only anticipate one disease, there is a potential that the death rate may rise as a result of the inability to foresee the other diseases. It is feasible to predict multiple diseases simultaneously using a multi disease model. Hence Users do not need to navigate numerous models in order to predict diseases. Time will be cut short, and there is a likelihood that fatality rates will go down because it can predict several diseases at once.
A Hybrid CNN and Gradient Boosting Framework for Spatio-Temporal Prediction Of Urban Crime Spots
(2026-02-09) Akshara Nair 1NH22CS014; Devarshi Khilariwal 1NH22CS065; Sohani Nagappa 1NH22CS256; Ashmith Maddala 1NH22CS273
Urban crime continues to pose significant challenges to public safety and policy enforcement, necessitating predictive systems that are both accurate and actionable. This research proposes a hybrid framework integrating Convolutional Neural Networks (CNNs) with Gradient Boasting techniques specifically CatBoost and XGBoost for the spatio-temporel prediction of urban crime hotspots. By leveraging historical crime dats enriched with spatial, temporal, and socio-demographic features, the system generates accurate forecasts of high-risk zones. CatBoost nancies categorical features such as crime types and locations, while XGBoost captures trends from numerical variables like population density and socioeconomic indicators. These models are ensemble-combined to improve predictive reliability. Additionally, CNNs are applied to heatmap-transformed data to extract spatial and temporal crime patterns.
A Hybrid Deep Learning Model for Multi Source Disaster Recognition Using ResNet-50 and Efficient Net
(2026-02-04) Kamalesh Navaneethakumar 1NH22CS103; Kamil Nissar 1NH22CS104; Faheera Kounain 1NH23CS404; Pooja Prakash Janagouda 1NH23CS411
The proposed system is designed to classify disasters using heterogeneous visual images. The framework integrated two state-of-the-heart convolutional neural networks: ResNet-50, which provides robust hierarchical feature extraction through residual learning, and EfficientNet, which achieves high accuracy with optimized computational efficiency using compound scaling.
A Lightweight Radar–Camera Fusion Framework For Real-Time Drone Detection And Trajectory Prediction
(2026-02-04) Nishanth Aluvala 1NH22CS147; P Krishna Kowshik Reddy 1NH22CS153; Levaka Umar Reddy 1NH22CS117; K Ashraf Ahmed 1NH22CS102
Unmanned aerial vehicles (UAVs), commonly known as drones, are increasingly used in civil, commercial, and security applications, leading to growing concerns related to airspace safety, privacy, and unauthorized intrusions. Traditional drone monitoring systems that rely on a single sensing modality often struggle under real-world conditions such as poor lighting, occlusions, cluttered backgrounds, or adverse weather, resulting in unreliable detection and tracking. To address these challenges, this project presents a lightweight radar–camera fusion framework for real-time drone detection and trajectory prediction. The system combines visual information from a monocular camera with motion-aware cues from radar data to achieve robust and consistent performance. A YOLOv8n model is employed for efficient drone detection from video frames, while a radar CNN processes range-Doppler maps to generate confidence scores that remain reliable even when visual cues degrade.
A Novel Approach for Monitoring Agricultural Production Process using Wireless Sensor Networks and Machine Learning
(NHCE, 2022) ADITYA ARABALE ; CHARANRAJ K R ; NISHANTH S B
The majority of nations rely heavily on agriculture. In India, agriculture directly supports more than half of the country's population. The yield of a given crop is influenced by a number of variables, including the climate, wind speed, soil quality, humidity, etc. The growth of a crop is impacted by these components' ongoing variability. The agricultural industry has benefited from technological advancement. In the agriculture industry, wireless sensor networks and crop yield prediction have had a significant impact. This study introduces a revolutionary Precision Farming method and demonstrates how sensor data can be effectively utilized. The Crop Yield Prediction model receives real-time input from the sensor-generated data. This strategy aids us in getting more accurate results. The project suggests a web application that, every split second, sends data from wireless sensors used in precision farming as an input to a crop production forecast model. Additionally, these sensed parameters from different users can be used as a training dataset. This method not only makes the most of the sensor data but also predicts crop yield with accuracy and promptness.
A NOVEL APPROACH FOR TRACKING THE SPREAD OF COVID-19 DISEASE AND DISCOVERING THE SYMPTOM PATTERNS OF COVID-19 PATIENTS USING ASSOCIATION RULE MINING”
(New Horizon College of Engineering, 2022) KEERTHY G-1NH18CS095 PRANAV NAG B -1NH18CS141
Covid 19 is a disease and has claimed millions of lives. Initial waves of Covid were spread in countries like China, and the USA. Later on, hundreds of countries started reporting cases of the virus causing panic among people. The virus is mutating itself and forming into newer ones. while the virus itself might not be that dangerous to all of them except a few old-age people or newborns. But when a person who has other problems like lung infection or many other diseases, contracts Covid then the person will be in trouble. there is constant research going on in this field with the use of the latest technologies as well. Even though the presence of current technologies like molecular tests known as PCR tests are used to detect the coronavirus genetic material. There are many other antigen tests as well. But all of them take a lot of time and there is a wide room for error like mixing up samples of different people or human error. So, the whole idea of this project is to reduce that error gap being created with the help of current technological methods. If the prediction is supposed to be true then the person can be sent for further detection of the virus. The model being built must be as accurate as possible since this is going to be used for medical purposes.
A Project Report on Kannada Kagunitha
(2026) Geethashree J M 1NH22CS079; Hema Pushpa J 1NH22CS088; Ashok G 1NH23CS402; Sunil Kumar D O 1NH23CS420
Educational Technology (EdTech) has advanced rapidly in the past decade, transforming how learners interact with academic content. Traditional classroom teaching has been augmented by digital learning platforms, self-paced modules, and interactive exercises that make education more accessible and enjoyable. This evolution has provided significant benefits for language learning, especially for early learners who respond well to visual, auditory, and interactive stimuli.
A RISK ANALYSIS TOOL FOR THE ASSESSMENT OF DIABETIC PAIENTS USING MACHINE LEARNING-BASED ALGORITHM
(NHCE, 2023) AAYET HASHMI : GOWTHAMIMD: MEGHANA CM
This project aims to address the growing concern of excessive screen time and its impact on well-being by leveraging generative artificial intelligence. The system involves user profiling, real-time monitoring, and adaptive learning to provide personalized recommendations for screen time moderation. Through a notification system and generative content creation, users receive timely alerts, motivational messages, and suggestions tailored to their preferences and health considerations. Privacy controls and regulatory ensure ethical data usage, while integration with various devices and platforms ensures a seamless and customized user experience. The project seeks to empower users with knowledge through educational content and foster healthy digital habits through continuous monitoring, evaluation, and user engagement strategies. Employing advanced machine learning models, behavioral analysis, and Generative Artificial Intelligence, our technical approach aims to address the pervasive issue of excessive screen time. With television and smartphones accounting for the majority of the excess screen time, over Indians surpassed recommended guidelines. This causes a negative impact on the well being of individuals, particularly the younger demographic. This major project in Computer Science and Engineering introduces an innovative solution leveraging Generative Artificial Intelligence (AI) to address and moderate digital screen time effectively.
A Secure File Sharing System
(2026-02-06) T Praneeth 1NH22CS229; Uday Kumar V 1NH22CS233; VVS Abhiram 1NH22CS238; Venkat Mohan Krishna V 1NH22CS241
The Secure File Sharing System is an application designed to provide a fast, reliable, and user-friendly method of transferring files between devices connected to the same local network. This system enables seamless communication between users by automatically discovering nearby devices, establishing secure connections, and allowing efficient file transfers without the need for internet access or external cloud services. The system consists of three major components: a Discovery Server for peer detection, a Backend Service for sending and receiving files, and a Frontend Interface that enables users to easily choose files, view active devices, and monitor transfer status.
A Smart IOT -ML Framework for Adaptive Crop and Fertilizer Recommendation in Precision Agriculture
(2026) Aishwarya K N 1NH22CS012; Bhagyashree M 1NH22CS043; Chaya K N 1NH22CS055; Deepthishree V 1NH22CS064
Agriculture has always been one of the most significant contributors to the global economy and human survival. It serves as the foundation for food security, economic development and employment generation, particularly in developing countries like India. However, in recent years, the agricultural sector has been facing several critical challenges such as climate change, soil degradation, water scarcity, unpredictable rainfall and the excessive use of fertilizers and pesticides. These challenges not only affect crop productivity but also threaten environmental sustainability and farmer livelihoods.
A SURVEY ON STOCK MOVEMENT PREDICTIONS
(New Horizon College of Engineering, 2022) K Aryanth Reddy -1NH18CS097 Ramiz Raja -1NH18CS155 Vamsi Kumar R -1NH18CS158
Accurate stock moment prediction of stock markets is tough and volatile owing to its unpredictable and non-linear increasingly complex, in order to meet the needs of the financial sector and financial market and raise the rate of return. The stock market is a trading platform where different individuals sell and buy stocks based on supply and demand. Stock market timing is difficult to anticipate because it is influenced by a variety of factors. Several research projects are carried out on a daily basis in order to predict accurate findings and extract the most value from them. Deep Learning algorithms and models have demonstrated accuracy in a variety of domains. It's difficult to predict how the stock market will perform and how it will move.Stock movement is tough because it is dependent on tens of thousands of investors.With better results, we implemented this project in a Jupyter notebook/Visual Studio.
A.C.C.O.L.A.D.E- Automatic Code Learner and Developer
(2016-07-13T06:07:39Z) Aishwarya, Warrier
Programming is hard. It requires a number of specialized skills and knowledge of the syntax of the particular programming language being used. Programmers need to know a number of different languages, that can vary in control structures, syntax, and standard libraries. As the programming languages evolve, it’ll be overwhelming to the programmer to learn as well as develop products that use these features in a short span of time. To overcome this problem, ACCOLADE is trained to learn frameworks of a programming language and automatically develop code for a given concept. The main idea of ACCOLADE is to learn code intelligently just like how our human brain learns and interprets information. ACCOLADE has two main purposes: CODE LEARNING and CODE DEVELOPMENT. In Code Learning phase it is given question/concept along with its associated code. In Code Development phase it develops code for the input question based on what it has learnt.
ABNORMAL EVENT DETECTION
(New Horizon College of Engineering, 2022) BHAGANNA HEBALI-1NH18CS040 CHALAPATHY V -1NH18CS047 N RAHUL REDDY -1NH18CS119
In the recent day’s surveillance cameras have increased drastically which can help in reducing crime but the surveillance cameras are not smart or intelligent to detect any abnormal event or violent event in the videos this makes the surveillance system always dependent on human evaluation for real time analysis it is not possible. hence an intelligent system that can trim out abnormal events such as violence, fighting, crowd- violence is of great importance. In the previous models the approach is to use audio to detect violence which is not possible always because some cameras don’t have microphone so, in this paper we are discuss all the present deep learning and machine learning algorithms for violence detection. and evaluate the various aspects of the algorithms with accuracy and precision.
Accessible Visual Information Retrieval(AVIR)
(New Horizon College of Engineering, 2024) YARTHA LIKITHA REDDY : 1NH20CS252 SIMI ROSHAN SHAM :1NH20CS280 GS JINITH KUMAR :1NH20CS306
Our method for Accessible Visual Information Retrieval (AVIR) stands at the forefront of technological innovation, seamlessly blending state-of-the-art tools like image. captioning, natural language processing (NLP), and human-computer interaction with the foundational principles of deep learning and machine learning. Our journey begins with a meticulous approach to dataset selection, ensuring diversity and richness, and then enriching them further with annotations meticulously gathered from invaluable user feedback. This iterative process lies at the heart of our commitment to inclusivity, guaranteeing that the resulting information caters comprehensively to the diverse needs of all users, including those with visual impairments. Whether it is providing alternative text for images or offering detailed audio descriptions, our technology enables meaningful engagement with visual information across various domains. In essence, AVIR transcends mere technological innovation; it represents a paradigm shift towards true inclusivity and empowerment, ensuring that every individual, regardless of visual ability, can fully participate in and benefit from the visual world around them.
Accident Evasion and Warning System
(NHCE, 2022)
Worldwide, a sizable fraction of traffic deaths take place every day. Developing automatic methods to recognise traffic accidents and reducing the amount of time it takes for first responders to get on the site after an accident are two efficient strategies to reduce the number of traffic deaths. The automated accident detection and alerting system that is fitted into modern vehicles is used. Although these techniques work, they are expensive, difficult to maintain, and not available in many vehicles. On the other hand, it has only recently been practical to use a smartphone to identify traffic accidents thanks to advancements in the processing speed and sensors used in cell phones. The bulk of smartphone-based accident detection systems use the vehicle's high speed (as determined by the smartphone's GPS receiver) and the G-Force number as their primary data points (extract from the smartphone accelerometer sensor). 90% of on-road collisions occur when the speed is low, according to multiple sources. Therefore, in addition to high-speed accident detection, low-speed accident detection was the main emphasis of our effort. Determining whether the user is inside the car or outside, strolling or gently running, presents the largest obstacle in averting a low-speed accident. In this work, the impact of this obstruction is lessened using a recommended approach that distinguishes between the speed variation of a low-speed vehicle and a walking or slowly moving human. The proposed system comprises two stages: prediction and detection. In the detection phase, both low- and high-speed vehicle accidents are found. Specific information, including images, videos, the location of the accident, and other details, are transmitted to the emergency responder during the notification phase and shortly after an accident is reported to ensure a speedy recovery. The technology performed incredibly well throughout testing in a simulation of real-world conditions. Keywords— Vehicle Tracking, Accident, detection, SMS notification, GPS, GSM
Accurate Evaluation System for Web Based Exam Management
(2026-02-06) Raushni P 1NH22CS176; Sanjana M 1NH22CS191; Shradhdha Sharad Kulkarni 1NH22CS202; Sinchana N 1NH22CS209
The Accurate Evaluation System is an application that aims at offering a smooth and effective system of conducting exams, managing students, and assisting the administration-related services within the learning institutions. This web-based system allows three roles, which are Admin, Student, and Teacher. Admins can control the users, assign teachers to the courses, and also check the profiles of the students. Students have the ability of enrolling, log-in, undertaking tests, and looking up their outcome and attendance. Educators will be able to test, tick attendance, and assess learners.
Achieving Flatness Selecting Honeywords From Existing User Passwords
(2017-08-16T06:58:10Z) Sushma, M; Supriya, G; Satwika, P; Sushmitha, Jain
Recently, proposed honey words (decoy passwords) to detect attacks against hashed password databases. For each user account, the legitimate password is stored with several honey words in order to sense impersonation. If honey words are selected properly, a cyber-attacker who steals a file of hashed passwords cannot be sure if it is the real password or a honey word for any account. Moreover, entering with a honey word to login will trigger an alarm notifying the administrator about a password file breach. At the expense of increasing the storage requirement by 20 times, the authors introduce a simple and effective solution to the detection of password file disclosure events. In this study, we scrutinize the honey word system and present some remarks to highlight possible weak points. Also, we suggest an alternative approach that selects the honey words from existing user passwords in the system in order to provide realistic honey words – a perfectly flat honey word generation method – and also to reduce storage cost of the honey word scheme.
Adaptive Control of Traffic Signalling System (ACTSS)
(2016-07-13T06:36:38Z) Abhishek, Y S S; Amar Bahadur, Thapa; Dinesh, Singh
Traffic congestion is a serious problem in developing and developed cities equally alike. Due to the ever growing number of vehicles on the road and the presence of the old infrastructure designed to handle fewer vehicles, it is not an easy challenge to tackle. This leads to problems like time loss, excessive pollution, etc. Modifying the existing infrastructure is a herculean task with lot of constraints and complications. The project is aimed to develop an adaptive traffic signalling system that avoids the need to upgrade the existing infrastructure and yet provides an efficient way to handle the traffic on the road. This is achieved by making the ordinary traffic system dynamic by considering real time data. Traffic density on the roads are considered and service time for respective roads are determined from it. There is a threshold value beyond which service time will not be allotted to prevent starvation of other roads. The scenarios that are handled by the system are, avoiding time wastage by providing service to an empty road, and avoiding the provision of excess time to a single road. The system is also aimed at detecting the presence of emergency vehicles, and if present, the regular service cycle is interrupted and the road with the emergency vehicle is serviced. Once the emergency vehicle is services, routine cycle will resume from the point where it was left. This adds a life-saving feature to the roads, which has not been implemented in most of the countries.

Browse

Browsing Project Report by Title

Results Per Page

Sort Options